Last chance! 50% off unlimited learning
Sale ends in
Hierarchical cluster analysis on a set of dissimilarities and methods for analyzing it.
hclust(d, method = "complete", members = NULL)# S3 method for hclust
plot(x, labels = NULL, hang = 0.1, check = TRUE,
axes = TRUE, frame.plot = FALSE, ann = TRUE,
main = "Cluster Dendrogram",
sub = NULL, xlab = NULL, ylab = "Height", …)
a dissimilarity structure as produced by dist
.
the agglomeration method to be used. This should
be (an unambiguous abbreviation of) one of
"ward.D"
, "ward.D2"
, "single"
, "complete"
,
"average"
(= UPGMA), "mcquitty"
(= WPGMA),
"median"
(= WPGMC) or "centroid"
(= UPGMC).
NULL
or a vector with length size of
d
. See the ‘Details’ section.
an object of the type produced by hclust
.
The fraction of the plot height by which labels should hang below the rest of the plot. A negative value will cause the labels to hang down from 0.
logical indicating if the x
object should be
checked for validity. This check is not necessary when x
is known to be valid such as when it is the direct result of
hclust()
. The default is check=TRUE
, as invalid
inputs may crash R due to memory violation in the internal C
plotting code.
A character vector of labels for the leaves of the
tree. By default the row names or row numbers of the original data are
used. If labels = FALSE
no labels at all are plotted.
logical flags as in plot.default
.
character strings for
title
. sub
and xlab
have a non-NULL
default when there's a tree$call
.
Further graphical arguments. E.g., cex
controls
the size of the labels (if plotted) in the same way as text
.
An object of class hclust which describes the tree produced by the clustering process. The object is a list with components:
an merge
describes the merging of clusters
at step merge
indicate agglomerations
of singletons, and positive entries indicate agglomerations
of non-singletons.
a set of method
for the particular agglomeration.
a vector giving the permutation of the original
observations suitable for plotting, in the sense that a cluster
plot using this ordering and matrix merge
will not have
crossings of the branches.
labels for each of the objects being clustered.
the call which produced the result.
the cluster method that has been used.
the distance that has been used to create d
(only returned if the distance object has a "method"
attribute).
There are print, plot and identify (see identify.hclust) methods and the rect.hclust() function for hclust objects.
This function performs a hierarchical cluster analysis
using a set of dissimilarities for the
A number of different clustering methods are provided. Ward's
minimum variance method aims at finding compact, spherical clusters.
The complete linkage method finds similar clusters. The
single linkage method (which is closely related to the minimal
spanning tree) adopts a ‘friends of friends’ clustering
strategy. The other methods can be regarded as aiming for clusters
with characteristics somewhere between the single and complete link
methods. Note however, that methods "median"
and
"centroid"
are not leading to a monotone distance
measure, or equivalently the resulting dendrograms can have so called
inversions or reversals which are hard to interpret,
but note the trichotomies in Legendre and Legendre (2012).
Two different algorithms are found in the literature for Ward clustering.
The one used by option "ward.D"
(equivalent to the only Ward
option "ward"
in R versions "ward.D2"
implements
that criterion (Murtagh and Legendre 2014). With the latter, the
dissimilarities are squared before cluster updating.
Note that agnes(*, method="ward")
corresponds
to hclust(*, "ward.D2")
.
If members != NULL
, then d
is taken to be a
dissimilarity matrix between clusters instead of dissimilarities
between singletons and members
gives the number of observations
per cluster. This way the hierarchical cluster algorithm can be
‘started in the middle of the dendrogram’, e.g., in order to
reconstruct the part of the tree above a cut (see examples).
Dissimilarities between clusters can be efficiently computed (i.e.,
without hclust
itself) only for a limited number of
distance/linkage combinations, the simplest one being squared
Euclidean distance and centroid linkage. In this case the
dissimilarities between the clusters are the squared Euclidean
distances between cluster means.
In hierarchical cluster displays, a decision is needed at each merge to
specify which subtree should go on the left and which on the right.
Since, for hclust
is to order the subtree so that
the tighter cluster is on the left (the last, i.e., most recent,
merge of the left subtree is at a lower value than the last
merge of the right subtree).
Single observations are the tightest clusters possible,
and merges involving two observations place them in order by their
observation sequence number.
Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988). The New S Language. Wadsworth & Brooks/Cole. (S version.)
Everitt, B. (1974). Cluster Analysis. London: Heinemann Educ. Books.
Hartigan, J.A. (1975). Clustering Algorithms. New York: Wiley.
Sneath, P. H. A. and R. R. Sokal (1973). Numerical Taxonomy. San Francisco: Freeman.
Anderberg, M. R. (1973). Cluster Analysis for Applications. Academic Press: New York.
Gordon, A. D. (1999). Classification. Second Edition. London: Chapman and Hall / CRC
Murtagh, F. (1985). “Multidimensional Clustering Algorithms”, in COMPSTAT Lectures 4. Wuerzburg: Physica-Verlag (for algorithmic details of algorithms used).
McQuitty, L.L. (1966). Similarity Analysis by Reciprocal Pairs for Discrete and Continuous Data. Educational and Psychological Measurement, 26, 825--831. 10.1177/001316446602600402.
Legendre, P. and L. Legendre (2012). Numerical Ecology, 3rd English ed. Amsterdam: Elsevier Science BV.
Murtagh, Fionn and Legendre, Pierre (2014). Ward's hierarchical agglomerative clustering method: which algorithms implement Ward's criterion? Journal of Classification, 31, 274--295. 10.1007/s00357-014-9161-z.
identify.hclust
, rect.hclust
,
cutree
, dendrogram
, kmeans
.
For the Lance--Williams formula and methods that apply it generally,
see agnes
from package cluster.
# NOT RUN {
require(graphics)
### Example 1: Violent crime rates by US state
hc <- hclust(dist(USArrests), "ave")
plot(hc)
plot(hc, hang = -1)
## Do the same with centroid clustering and *squared* Euclidean distance,
## cut the tree into ten clusters and reconstruct the upper part of the
## tree from the cluster centers.
hc <- hclust(dist(USArrests)^2, "cen")
memb <- cutree(hc, k = 10)
cent <- NULL
for(k in 1:10){
cent <- rbind(cent, colMeans(USArrests[memb == k, , drop = FALSE]))
}
hc1 <- hclust(dist(cent)^2, method = "cen", members = table(memb))
opar <- par(mfrow = c(1, 2))
plot(hc, labels = FALSE, hang = -1, main = "Original Tree")
plot(hc1, labels = FALSE, hang = -1, main = "Re-start from 10 clusters")
par(opar)
### Example 2: Straight-line distances among 10 US cities
## Compare the results of algorithms "ward.D" and "ward.D2"
mds2 <- -cmdscale(UScitiesD)
plot(mds2, type="n", axes=FALSE, ann=FALSE)
text(mds2, labels=rownames(mds2), xpd = NA)
hcity.D <- hclust(UScitiesD, "ward.D") # "wrong"
hcity.D2 <- hclust(UScitiesD, "ward.D2")
opar <- par(mfrow = c(1, 2))
plot(hcity.D, hang=-1)
plot(hcity.D2, hang=-1)
par(opar)
# }
Run the code above in your browser using DataLab